Critical MaxKB AI Agent Flaw Lets Prompt Injection Execute System Commands
A critical vulnerability in the MaxKB AI knowledge-base platform lets attackers execute OS commands via prompt injection, using the sandbox shell backend and elevated privileges; it can lead to remote code execution.…
A critical vulnerability in MaxKB that allows shell commands via prompt injection, with CVEs CWE-78 (OS command injection), CWE-250 (execution with unnecessary privileges), and CWE-749 (exposure of dangerous functionality).
- - CVE-2026-77521, maxkb sandbox, shell command injection, OS command injection, user privileges, elevated permissions
- - MLM-001, root shell bypass, unauthenticated attackers, multi-tenant deployments
- ATT&CK
- T1566.001 Spearphishing AttachmentT1190 Exploit Public-Facing ApplicationT1486 Data Encrypted for Impact
- Products
- Threat actors
- LockBit
- Malware
- MLM-001
- Victims
- CryptocurrencyDNS service
- AI models
- MaxKB
From the original article
3 min read at gbhackers.com
A critical vulnerability in the MaxKB AI knowledge-base platform could let attackers exploit prompt injection and run operating system commands on vulnerable deployments, including directly on the underlying host in some configurations. This flaw, tracked as CVE-2026-77521 and GHSA-f36j-f34j-h3rx, affects MaxKB versions up to and including 2. 10.
By Divya · gbhackers.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.